Alcoholic hepatitis is a serious consequene of excessive alcohol consumption and may result in death during the active stages of the disease or in the later development of cirrhosis. We are investigating the effects of prednisolone versus placebo therapy in patients with severe alcoholic hepatitis to determine if such therapy influences early mortality rate of resolution of liver function test abnormalities and instance of complications. The studies are performed in a randomized, double-blind, one-month trial to take place in four centers. Admission to the study shall be limited to those patients with severe alcoholic hepatitis as evidenced by encephalopathy early in their course and/or marked prolongation of prothrombin time and elevation of serum bilirubin as defined by a previously determined discriminant function (4.6 x the prolongation of the prothrombin time in seconds plus the height of the serum bilirubin). Patients shall be entered in the study when the discriminant function value exceeds 32 with the score taken on the fourth to sixth hospital day. Mortality in the two treatment groups will be compared using standard methods of survival analysis which take into account length of survival as opposed to crude survival rates. Covariants shown to be important in determining survival will be included (adjusted for) in this analysis. The usefulness of the discriminant function (above) in predicting risk of death will be prospectively evaluated and compared to a logistic discriminant function to be developed during the study. Effects of therapy on a wide variety of laboratory tests will be studied by standard statistical methods.